Re: current limit on 48 VDC
OK, my 2 cents, and I apologize for those who will get this twice.
I believe that I started the whole conversation about PTCs for current
limitation, and I'd like to get one point straight. My point of reference was
specifically trying to discuss the limitations of a "FAULT" state in a high
level PSE state machine. Specifically, I believe that in most cases, a PSE
should exit a "FAULT" state once an external physical fault is removed (such as
a short in the wires is fixed or a crowbarred PD is unplugged, etc.) and return
to normal operation without requiring manual intervention. I believe that
specifying that a PSE must automatically recover is a reasonable and good thing
to do in our standard, it allows users to know what behavior to expect from any
device that adheres to the spec. To that end I discussed a timer in relation to
how long it was acceptable to remain in the "FAULT" state after the fault was
cleared (at least that was what I intended to discuss, perhaps not eloquently
enough though.) I have suggested that the recovery time of a PTC is one of the
things that should be considered when setting the timer limit (if there is one.)
I am not going to suggest that a PTC is the perfect device for our power
limitation application, nor am I going to suggest that no one can design a
circuit that contains a PTC that can't meet the strictest cut off current or
shut down time we could come up with. I am only suggesting that if we allow a
PSE to stay in a "FAULT" state for a limited period of time after a fault
condition is removed, that perhaps it is wise to allow for the recovery time of
PTCs as they are an example of a device that requires a recovery time that I
know to have been used in a fault isolation circuit.
With all that said, I think the exact choice of power protection technology is
an implementation issue, that we would do a disservice to the industry to
specify one technology while excluding all others. The discussion about what
technology to use to provide the current cut off is something that the PSE
designers should have with their power protection technology provider(s) and
does not need to be decided once by the committee. However, if we set limits
for such things as fault recovery timers at certain values, we may
unintentionally exclude some technology that may be desirable in some
applications.